Confusion and agitation in the elderly is a crucial nursing problem. Limitations and concerns related to the current management of these behaviors indicate a critical need to investigate alternative nursing interventions. The proposed study will test Gerdner's Explanatory Theory to investigate individualized music as one such intervention. A quasi- experimental repeated measures pre-test post-test cross over design will be used to compare the immediate, 30 minute residual, and weekly cumulative effects of individualized music to classical 'relaxation' music (relative to baseline) on the frequency of agitated behaviors in elderly persons with ADRD. Independent variables will be measured by using the Cohen-Mansfield Agitation Inventory and a modified version thereof. A comparison will also be made of the differential effectiveness of each intervention in relation to the degree of the subject's cognitive impairment, based on the Global Deterioration Scale. A randomized sample of 40 subjects will be selected from four long term care facilities in Iowa. Baseline data will be collected during the initial three weeks. The time of intervention will be individualized based on application of the Progressively Lowered Stress Threshold Model and findings from the Temporal Patterning Assessment of Agitation. The selection of individualized music will be based on findings from the Modified Hartsock Music Preference Questionnaire. Group A will receive individualized music for six weeks, followed by a two week 'washout' period to nullify possible cumulative effects. Classical 'relaxation' music will then be presented for six weeks. Group B will receive the same protocol but in reversed order. Music interventions will be presented to each subject two times per week. Analysis of data will be conducted using a repeated measures analysis of variance (ANOVA) and a graphic display of comparative mean frequencies of agitated behaviors.